Inspection penetrants are used to find surface cracks in all types of insoluble materials, but primarily in nonmagnetic metals, such as aluminum, high nickel alloy, stainless steel and titanium. These metals predominate in the aerospace industry. On many aerospace parts, such as rotating discs of turbine engines, even the most microscopic cracks must be found, as these microscopic cracks will propagate under the stresses and loads placed on them during operation. Propagation can be extremely rapid and part failure sudden and catastrophic.
Inspection of such critical parts for surface cracks, whether newly manufactured or parts that have been in service, is accomplished with fluorescent penetrants. The process consists of application of fluorescent penetrant, an oily-like fluid charged with fluorescent dyestuff, to the surface of the part by immersion, brushing or spraying; allowing the fluorescent penetrant to remain on the surface for sufficient time to enter microscopic surface openings by capillary action; removing the penetrant that did not enter cracks and surface flaws from the surface by washing with a pressure rinse of plain water or with the aid of a detergent or emulsifying agent. The part is then dried in an oven and typically a developing agent is applied to bring the crack entrapped fluorescent penetrant back to the surface so it may be seen as a glowing line under ultraviolet light (black light) in a darkened room.
The fluorescent penetrant system must function properly or cracks will go undetected. These cracks can propagate into catastrophic failures. For example, if the penetrant has become contaminated with different substances, its fluorescing properties will be diminished. If the mechanism for applying the developing agent is malfunctioning, the flaws will not be clearly identifiable under the ultraviolet light. There are other causes for failure such as rinse water with a too-high temperature and a too-high concentration of detergent which will prematurely remove penetrant from cracks.
To defend against processing critical aerospace and turbine engine parts in a malfunctioning fluorescent penetrant system, a general requirement is to prove the system each day before it is used by processing a test panel with known defects. If, after processing, these known defects are not displayed as anticipated, i.e. with the same completeness and brilliance as in previous tests, this alerts the operator of the need to check the system for a malfunction.
Probably the best known panel for this purpose, and most widely used, is the PSM-5 panel manufactured by Sherwin Incorporated, South Gate, Calif., to Pratt and Whitney drawing TAM 146040, which drawing is dated April 1975. The PSM-5 panel is a 4xc3x976xe2x80x3 piece of stainless steel, thickness 0.090xe2x80x3 with a strip of hard chrome plate running lengthwise down one side. The thickness of this chrome plate is 0.003xe2x80x3 or greater. As hard chrome plating is applied electrolytically, its thickness will vary over the surface with heavier coating to be anticipated at the edges. Five cracks of varying magnitude, evenly spaced, are induced by exerted pressure opposite the hard chrome strip with a Brinell hardness test instrument. The balance of the front of this test piece and adjacent the chrome strip is a rough area obtained by grit blasting with aluminum oxide or other media.
The most difficult task in manufacturing the PSM-5 panel is in the formation of the smallest crack, a crack diameter in the range of 0.015xe2x80x3 to 0.030xe2x80x3. This small crack, necessary to verify the system""s ability to find the truly microscopic crack, is difficult to produce, as there is no room for the slightest error in the plating composition or in the pressure exerted by the hardness tester. The crack is formed by pressure exerted by a hard round ball on the side opposite to the plating. The plating side is backed up against a selected surface. The pressure is measured by weight, pounds or kilograms. Although the hardness test equipment includes an instrument to indicate in kilograms the weight force, there is a lag in indication and this equipment is not sufficiently precise to give controlled crack formation in the area of small cracks whose detection is required by today""s advanced industry.
The method of inducing cracks with a hardness tester has an inherent deficiency. If the induced crack fails to meet its specification, it is not possible to review and remeasure to verify if the prescribed pressure was applied. The indentation left by the hardness tester is not measurable with ordinary tools.
As the aerospace industry continues to reduce the weight of their vehicles while at the same time demanding higher performance and placing greater stress loads on the components, the need for the penetrant process capability to locate smaller surface flaws becomes more critical. The need for a test piece that verifies the inspection system""s capability to meet these more demanding specification requirements is obvious.
No two PSM-5 panels are exactly alike. They are produced individually. Further, the cracking method cannot be precisely controlled. It is not possible to hold two PSM-5 panels side-by-side and expect to see equivalency in the crack patterns. The PSM-5 panel cannot be used to reliably compare relative sensitivity between two different penetrants because no two panels are equivalent. One cannot expect to process one panel with new penetrant and another with in-use penetrant and obtain a meaningful comparison. This is due in part to the fact that the panels are produced separately and not as one piece. It is also due to the method of cracking, applying pressure until cracking occurs.
Recognizing this deficiency in the PSM-5 panel, governing agencies now require the user of the PSM-5 panel to photograph the panel when first processed with unused penetrant materials in the laboratory and, then to use this photograph to compare results obtained when the panel is subsequently processed through the production penetrant system on a daily basis. Although this has some utility, it is not truly satisfactory, since, in order to take photographs under ultraviolet light, time exposures are required and the photographed fluorescent crack indications will vary in size and definition with exposure time, as well as with film negative and printing paper and technique. Further the photograph must be viewed under white light, as the crack indications in the photograph do not fluoresce. The panel itself must be read under ultraviolet light. Such indications cannot be viewed and compared to actual fluorescing indications in the darkened inspection booth where the lighting is ultraviolet and be meaningful.
But this recently imposed requirement of a photographic comparison in such specifications as ASTM E 1417, despite its inadequacies, is evidence of the need of a reference point when interpreting the panel""s results.
Also, although the PSM-5 cracks may be small, e.g. 0.015xe2x80x3 of an inch in diameter, their depth is the depth of the plating which is 0.003xe2x80x3 or greater. The nature of the hard chrome requires a plating thickness of close to 0.003xe2x80x3 to crack with a pressure load, otherwise, it stretches until it splits uncontrolled. A crack 0.003xe2x80x3 or more in depth retains considerable penetrant and, therefore, does not reveal abusive over-washing and over-heating in processing as readily as it should. Further, a crack with this depth tends to retain penetrant even when subjected to extensive cleaning between tests. Such retained penetrant from a previous test leads to erroneous conclusions on subsequent tests. ASTM E 1417 stresses the need for adequate cleaning of the known defect standard between tests.
To meet today""s need, the brittle coating must lend itself to controlled cracking in thin coatings, less than 0.002xe2x80x3. Thin coats of chrome plating, such as 0.001xe2x80x3, do not crack uniformally even though pressure exerted is uniform. Although there may be a place for chrome plating, metal conversion coatings, silicate and other brittle coatings, and these coatings lend themselves to the controlled method of cracking that I discovered, my favorite brittle coating is nickel plate, either electrolytic or electroless nickel type, because a thin coat of nickel plate lends itself to controlled cracking. Cracks can be induced with my invention in nickel plate in a coating as thin as 0.00025xe2x80x3 in a controlled manner.
Although we have fabricated panels with plating thickness of 0.00025xe2x80x3 which have proven to be a practical tool, typically, we have found the plating thickness of 0.001xe2x80x3 the most useful. Cracks of this depth duplicate the small shallow cracks which are the object of today""s most exacting inspections. The penetrant retained in a crack reservoir 0.001xe2x80x3 deep with a diameter of 0.015xe2x80x3 is sufficiently minuscule for the entrapped penetrant to be affected deleteriously by over-washing, over-heating and inadequate developer application. Abusive processing or a substandard penetrant material is more readily apparent with a test piece with shallow, 0.001xe2x80x3 cracks than when crack depth of the test piece is 0.003xe2x80x3 or greater.
An added advantage of the thinner coating is the ease of cleaning the panel and freeing the crack reservoirs of materials deposited during previous use. A 15 minute soak in alcohol is all that is required to clear the reservoirs whereas a lengthy bath in an ultrasonic cleaning tank charged with a chlorinated solvent often has to be repeated to clear the crack reservoirs of the PSM-5 panel.
Like the PSM-5 test piece, my favorite substrate is stainless steel, as it is rugged in construction, not subject to corrosion, and withstands the rough handling when sent through the penetrant system.
There are other test pieces used to evaluate inspection penetrants, such as thermally cracked aluminum blocks and xe2x80x9cnickel-chromexe2x80x9d test panels. These panels are not suitable replacements for the PSM-5 type panel and are not intended for use in evaluating the functioning of an inspection penetrant system.
The nickel-chrome test panel for the sake of clarification should be described, so it will be understood that it does not compete with our invention. It is known in the industry as the xe2x80x9ctwin nickel-chrome panel.xe2x80x9d This panel""s substrate is brass, subject to corrosion. The plating is brittle nickel with a flash of chrome. Panels usually measure about 1xc2xcxc3x974xe2x80x3 and are 0.06 thick. The plating will vary in thickness from as thin as 0.0002xe2x80x3 (5 xcexcm) to 0.002xe2x80x3 (50 xcexcm).
The panels are bent over curved anvils to induce cracking and then straightened, as shown in U.S. Pat. No. 4,610,157. The main difficulty with the xe2x80x9ctwin nickel-chrome panelxe2x80x9d is the form of the cracks, straight cracks, running laterally from one edge of the test piece to the other edge. The cracks"" geometries are open troughs. Penetrant that enters this type of cracking is easily flushed through the open trench or trough. While it is possible to compare the visibility of two penetrants by judicious removal of surface penetrant, it is not possible to process these test pieces in the work environment. The penetrant flushes from the xe2x80x9ctroughsxe2x80x9d too readily. These test pieces have a limited function. For this and other reasons, they are not a practical tool for monitoring a penetrant system. The xe2x80x9ctwin nickel-chrome panelxe2x80x9d is primarily a laboratory tool. Another patent which shows a test panel process is U.S. Pat. No. 4,078,417.
The nickel-chrome test panel is normally produced as a single panel which is later sheared into matching panels. It can be used for side-by-side comparisons of different penetrant materials in the laboratory. It is not sufficiently rugged to be used to measure the capability of a production penetrant system and it gives erroneous data relative to the wash cycle.
The present continuation in part application is concerned with the means to further control the magnitude of the induced cracks made in the test piece. The objective of the improved process is to manufacture test pieces with precise crack sizes, both in crack depth and crack length. The use of the penetration method described in the present application has made possible precision induced cracking in the production of virtual twin test pieces where a single brittle coating specimen is shared into equivalent sections as the brittle plating coating is applied at the same time on both sections. However, it has been difficult or impossible to extend this control when the plating differs, which is the case with specimens plated at different times. The present improvement makes possible such depth and length control by confining the cracking force to a predetermined limited area by weakening the substrate to such a predetermined limited area.
It was discovered that by drilling holes part way through the substrate from the back of the substrate toward the plated side, cracks could be induced in a much more controlled area, namely, the area equal to the diameter of the hole which has been drilled part way through the substrate. The tool used to produce cracking described in the present application is inserted in the incompletely drilled holes and forced against the bottom of the hole an amount sufficient to create cracks in the plated surface on the plated side of the test plate. The result was a highly controlled crack depth and length which did not require the plating of other panels to be done at the same time.
The present invention is for a method of forming controlled cracks in a predetermined area of brittle plating on a plated surface of a test plate which comprises a substrate which has a plated area on a plated side and a back side. One or more holes are drilled part way through the substrate from the back side of the test plate opposite a plated area of the test plate to create a cavity having a cavity diameter and a cavity bottom. The point of a hardened tip is inserted into the cavity and forced against the cavity bottom with sufficient force to create a deformation of a portion of the plated area beneath the cavity bottom. This forms a plurality of cracks in the plated area limited to an area no larger then the cavity diameter. Preferably, holes of different sizes are drilled in the test plate to provide cracks of different depth and length. The holes are preferably drilled more than half way through the substrate.